Incorporation of microbiological and molecular methods in HACCP monitoring scheme of molds and yeasts in a Greek dairy plant: A case study

Transcription

1 Available online at Procedia Food Science 1 (2011) th International Congress on Engineering and Food (ICEF11) Incorporation of microbiological and molecular methods in HACCP monitoring scheme of molds and yeasts in a Greek dairy plant: A case study Evagelos Beletsiotis a, Dimitris Ghikas a*, Kelly Kalantzi a, a DELTA FOODS S.A., Agios Stefanos, Athens 14565, Greece Abstract Yogurt and yogurt based products are widely consumed globally within European countries. The main microbiological spoilage problems occur from acid tolerant yeasts and molds. The prevention and /or reduction of such contamination is a major objective of HACCP. One of the prerequisite control points is the microbiological air quality of the facility. A one year prospective study of fungal air contamination was conducted in one outdoor and two indoor areas, in a Greek dairy plant. Air was sampled with a portable air system impactor and identified mainly through classical microbiological methods and molecular typing techniques. During the period of survey, the mean viable fungal load was CFU/m 3 in outdoor air samples and 69.8 CFU/m 3 and CFU/m 3 in samples from the two indoor locations. The three dominant fungal genus recovered were Cladosporium spp., Penicillium spp. and yeasts. Seasonal variations observed in all locations were examined and corrective actions were applied to reduce the fungal load, in one of the interior areas. HEPA air filters were installed and a twenty time reduction (from 69.8 to 3.5 CFU/m 3 ) was achieved concerning the total fungal load. Implementation of a constant monitoring of the air quality and the recognition, as a critical control point (CCP) led to lower fungal air load, furthermore enhancing the microbiological quality and safety of the products Published by Elsevier B.V. Open access under CC BY-NC-ND license. Selection and/or peer-review under responsibility of 11th International Congress on Engineering and Food (ICEF 11) Executive Committee. Keywords: airborne fungi; yogurt; dairy; Penicillium; Cladosporium 1. Introduction Yogurt is a fermented low acid dairy product made mainly with the use of bacterial cultures of Streptococcus thermophilus and Lactobacillus delbrueckii subsp. bulgaricus. Because of its high nutrient value, it is widely consumed over the world. It is a relatively stable physicochemical product and one of its quality problems is the microbiological spoilage occurring from molds and yeasts [1]. Contamination * Corresponding author. Tel.: ; fax: address: dimghi@vivartia.com X 2011 Published by Elsevier B.V. Open access under CC BY-NC-ND license. Selection and/or peer-review under responsibility of 11th International Congress on Engineering and Food (ICEF 11) Executive Committee. doi: /j.profoo

2 1052 Evagelos Beletsiotis et al. / Procedia Food Science 1 (2011) with fungal spores and yeast fruiting bodies could occur during production, after pasteurization and during the packaging. Major contaminant problem constitutes fungi belonging mainly in the phyla Ascomycota and Zygomycota, having the ability to grow in low acid, low temperature environments and could cause optical and sensory deterioration of the product. More specific, genera of Aspergillus spp., Penicillium spp., Fusarium spp., Alternaria spp. and different yeasts genera, like Candida spp. (belonging to Ascomycota) and Mucor spp., Rhizopus spp. (Zygomycota) are the most frequent for product contamination. As sources of contamination have been recognized packaging material, adding ingredients, water and air indoor the manufacturing facility [2, 3]. The prevention and/ or reduction of such contamination sources is a major objective of HACCP (Hazard Analysis Critical Control Point) systems with the adoption of GMPs (Good Manufactured Practices) in dairy plants. Most of the practices focus on the maintenance of a proper level of cleaning in the equipment and the plant interior and only few recognize the microbial quality of air, as a source of contamination, although air has been established as a source of fungal contamination on dairy industry [4, 5]. Introduction of an air monitoring scheme, recording the microbiological load as an HACCP procedure, has been applied in a number of food companies and hospitals. Such a monitoring scheme was applied on a Greek dairy industry, collecting data on a weekly basis. The microbial quality and quantity of air determined using mycological and molecular techniques, mainly focused on the spoilage genera of molds, using viable culture techniques. Microbiological data from a twelve month period were obtained, and analyzed in order to determine the impact of the fungal air load and the occurrence of specific genera of fungi in the product quality. Fungal seasonality was studied, critical for contamination areas assigned and preventive measures to minimize the risk were taken. In order to minimise the fungal air load in the production line, High Efficiency Particulate Air (HEPA) filters were installed and the resulting microbiological air quality was studied and assessed. 2. Materials & Methods 2.1. Plant and indoor areas The study conducted in a dairy plant located in Attica, near a semi agricultural area, 25 Km North of the centre of Athens (Greece). The interior of the facility is separated into process, filling, packaging and paletization areas. Air samplings were collected from eight different indoor locations IN1 to IN8 areas and two different outdoor areas (OUT1-OUT2) Air sampling protocol Duration and frequency Air sampling was carried out for a period of twelve months from September 2004 to August Concerning samples, the analyses presented in this study centred in two indoor areas (IN1 and IN5) and one outdoor area (OUT1). IN1 located in the main filling area while IN5 was in the area of paletization of the ready product. OUT1 sampling area located at the exterior environment and in proximity to the IN5 area. At the indoor locations, samples were collected once a week before noon, from one meter height, while the outdoor sampling was made twice a week Sampling Method Method of choice to estimate the number and species of fungi was the viable sampling method. Air samples were collected using the mobile spore trap Air IDEAL (BioMerieux) for agar plates. Air passes through a 256 holes grid over a 90mm diameter petri dish filled with Malt Extract Agar (MEA, Merck). Air intake was set at 100L/min. Collection time for indoor and outdoor samples were 1 min per petri dish (100 l of air). At IN1 location after the installation of HEPA filters at May 2005, the collection time was

3 Evagelos Beletsiotis et al. / Procedia Food Science 1 (2011) readjusted to 10 min per petri dish (1000 l of air). The most probable number of colonies recovered on the air sample plates were calculated with the instructions of manufacturer s manual. The results (concentration of airborne fungi) were expressed as colony formation unit per cubic meter Microbiological-Mycological Methods MEA petri dishes were incubated at 25 o C for 5 days (and in some cases up to two weeks). After incubation period passed, petri dishes were photographed and fungal species were identified, macroscopically and microscopically, according to their morphological characteristics (colour, growth) and the formation of their reproductive apparatus [6, 7]. Newly occurring species were further molecular analyzed and catalogued in order to build an in house myco-bank containing data like species characterization, site of occurrence, frequency and seasonality Molecular tools A number of strains were further analyzed/ identified in species level using molecular techniques. 100 mg of single colony well grown fungi were used for DNA isolation with Ultra-PrepMan kit (Applied Biosystems - ABI). Molecular identification was performed using either D2LSU rdna kit (ABI). PCR and sequencing reactions were made in a 9800 Fast PCR (ABI) according to manufacturer instructions. Sequencing reactions were analyzed in a 310 sequencer (ABI) Statistical analyses Because of asymmetric data distribution, nonparametric tests were used Mann-Whitney U-test setting the level of significance at Analyses were performed using Excel 2000 (Microsoft Corp.)Files should be in MS Word format only and should be formatted for direct printing. Figures and tables should be embedded and not supplied separately. Please make sure that you use as much as possible normal fonts in your documents. Special fonts, such as fonts used in the Far East (Japanese, Chinese, Korean, etc.) may cause problems during processing. To avoid unnecessary errors you are strongly advised to use the spellchecker function of MS Word. Follow this order when typing manuscripts: Title, Authors, Affiliations, Abstract, Keywords, Main text (including figures and tables), Acknowledgements, References, Appendix. Collate acknowledgements in a separate section at the end of the article and do not include them on the title page, as a footnote to the title or otherwise. 3. Results & Discussion 3.1. Airborne fungal concentrations During the 12 months period of surveillance, a total number of 104 air samples were collected in total, for the indoor locations (IN1 and IN5) and 92 samples for the outdoor location (OUT1) (Table 1). The outdoor mean concentration of fungal load was calculated at CFU/m 3. Same fungal loads have been recorded in city of Athens (Greece) using viable sampling method with annual mean value for recovered fungi at CFU/m 3 [8]. In Paris (France) at 2003 the airborne fungal concentration measured was below 500 CFU/m 3, while 300 Km away, in Dijon (France) the mean value of fungal load was CFU/m 3 at the same period of time, almost three times lower than the measurements of our study [9]. Geographical and climate variations, differences in sampling and measurement techniques like air flow rate, sampling duration, cultivation medium used, are among the factors that could influence the results.

4 1054 Evagelos Beletsiotis et al. / Procedia Food Science 1 (2011) The fungal load was lower in the interior areas with 93.6 CFU/m 3 for IN1 and CFU/m 3 for IN5 respectively. The differences in fungal load between the two interior locations were mainly due to the proximity of IN5 to the exterior environment and the plant working practices. More specific, periodically during the day, doors were opened for product and vehicle movement allowed exterior air to enter the facility s interior. APHA (American Public Health Association) recommends that standards for aerobic plate count in the air of food processing areas could be at 90 CFU/m 3 evaluated by the air sampler technique [10]. Kang and Frank [4] recorded CFU/m 3 for yeasts and molds, according to food processing areas, while in studies of different processing areas in dairy plants, the fungal load recorded was between CFU/m 3 [11]; [12]. There are no data available for indoor fungal load occurrence and genera distribution in Greek industries. Data exist only for hospital departments [13]. The airborne fungal loads measured in the IN1 area, after the HEPA filters installation (3.5 cfu/m 3 ), were comparable to the Intensive Care Unit and Hematology Department of Athens s and Thessaloniki s hospitals with values ranged from 2.4 CFU/m 3 to 9.6 CFU/m 3 [13], indicating a very good microbiological quality of air. Studies in the airborne fungal load and profile in the interior of free of water damage and fungal growth - healthy houses in different regions of USA showed values ranged between CFU/m 3, while the median concentration of outdoor fungal load ranged between CFU/m 3 [14]; [15]. Mean values measured at both interior areas (IN1 and IN5) were comparable to these mentioned above, indicating that the air quality inside the facility was comparable to that of a healthy house interior. Table 1. Concentrations of fungi collected in one outdoor area and in two indoor areas of the plant facility during a period of one year. Location Number of samples Mean value of viable fungi in CFU/m 3 Positive samples Outdoor - OUT Indoor area 1- IN1 38 (a) Indoor area 1- IN1 14 (b) Indoor area 5- IN (a): before HEPA installation, (b): after HEPA installation 3.2. Fungal genera and annually isolation frequencies Table 2 shows the overall frequencies and the corresponding fungal load of the fungi isolated in outdoor air and in two interior plant areas in annually basis. The sampling method and growth medium allowed the maximum recovery of filamentous fungi and yeasts [8]. The identified fungi belonged to genera Penicillium, Cladosporium, Aspergillus, Alternaria, Fusarium, yeasts and other filamentous fungi. As non sporulating fungi (N.S.F.) designated those which showed no spore formation after two weeks of incubation. In outdoor air samples, higher concentration was observed for Cladosporium sp. (62%), followed by Penicillium sp. (8.8%) and Alternaria sp. (7.9%). An additional 12.8% of the total load represented different genera of yeasts, while the other 8.4% in total corresponded to Fusarium sp., Aspergillus sp., Trichoderma sp., Paecilomyces sp., Geotrichum sp., Aureobasidium sp., Ulocladium sp. (Phylum Ascomycota), Mucor sp. (Phylum Zygomycota), N.S.F and other fungi. Indoor air samples from IN1area contained mainly Penicillium sp. (61.5%) and Cladosporium sp. (14%). Low levels of yeasts (5.7%), Fusarium sp. (5.4%) and Aspergillus sp., (2.5%) were also present, while the other 9.4% consisted of Mucor sp., Trichoderma sp., Geotrichum sp., Trichosporon sp., N.S.F and other fungi. Same distribution with differences in occurrence was obtained from IN5 location. More specific, Penicillium sp. and Cladosporium sp. with 49.8% and 26.1% respectively, where the dominant genera isolated, followed by yeasts (11.9%), Alternaria sp. (2.7%) and Fusarium sp. (2.7%). The other

5 Evagelos Beletsiotis et al. / Procedia Food Science 1 (2011) % consisted of Mucor sp., Trichoderma sp., Geotrichum sp., Trichosporon sp., Aureobasidium sp. and other fungi (Table 2). Table 2. Fungi identified in air from outdoors and from both indoor locations during the study period. Mean Fungal Load () and entage () of total appeared. S.D: Standard deviation. Indoor 1 (IN1) Indoor 2 (IN5) Outdoor (OUT1) Fungi (genera) SD (+/-) Cladosporium sp Penicillium sp SD (+/-) SD (+/-) Aspergillus sp Alternaria sp Ulocladium sp Fusarium sp Mucor sp Rhizopus sp Paecilomyces sp Trichoderma sp Geotrichum sp Aureobasidium sp Yeasts sp Trichosporon sp N.S.F Other Concerning percentages of fungi isolated in outdoor air during sampling period, they are in accordance with previous study with worth mentioned differences [8]. For instance, Cladosporium sp. percentage was almost twice as higher than that, isolated at 1998, while percentages of Penicillium sp. and Alternaria sp. were comparable [8]. These differences in fungal distribution could be due to the large chronicle distance (10 years) between the two seasonal measurements and different locations, which reflects different microclimatic conditions (humidity, temperature, dust load etc). On the other hand, our results are in good accordance with results obtained from Dijon (France) where the occurrence (percentage) of fungal genera in outdoor air was similar for Cladosporium sp., Penicillium sp. and Alternaria sp. Indoor results revealed a different fungal genera distribution. Penicillium spp. were the predominant species isolated followed by Cladosporium sp., Fusarium sp. and yeasts. Studies in USA houses [14], showed that the occurrence pattern of airborne fungal genera maintained indoor when compared with outdoor air fungal load. However in our study, higher percentage of Penicillium sp., and Fusarium sp. were observed indoor, indicating species adaptation inside the plant microenvironment (occurrence of fungal sporia on walls, floors, surfaces). Same fungal distribution pattern existed inside hospital facilities in Dijon (France) [9]. Our results revealed high percentages of yeasts genera. Main cause was the high air humidity percentage, related to manufacturing practices. Differences in fungal air load of Cladosporium

6 1056 Evagelos Beletsiotis et al. / Procedia Food Science 1 (2011) sp. and Alternaria sp., in the two indoor areas, indicate microbiological transfer between exterior and interior environment at IN5 area Table Fungal seasonal variation and occurrence Table 3 shows the seasonal variation of the colony count for the three environments studied. In outdoor air, maximum fungal load was enumerated in summer (537.7 CFU/m 3 ) and autumn (392.2 CFU/m 3 ) with Cladosporium sp. to be the predominant genus with CFU/m 3 and CFU/m 3 respectively. Lower Cladosporium sp. counts were observed in winter (22%, 53.3 CFU/m 3 ) while in the same period yeast species reached the highest mean value of CFU/m 3 (45.2%). Same seasonal fluctuation pattern were observed for Alternaria sp. Furthermore, in winter, the total airborne fungi recovered had the lowest counts (235.4 CFU/m 3 ) compared to other seasons. Penicillium species reached the highest percentage during winter (13.6%), although the maximum fungal load was obtained in fall (38.4 CFU/m 3 ) and the lowest during summer (6.1%). Same distribution pattern was observed for Fusarium sp. (13% in winter and 0.8% in summer). Yeasts were recovered in highest values during winter (45.2%, CFU/m 3 ). Higher outdoor concentrations of total fungal load in the summer and autumn could reflect higher temperatures and humidity, resulting to higher microbiological air load. Cladosporium is the most dominant spore forming fungi, occurring outdoors in many cities around the world [8]. Occurrence of fungal genera, belonging to Penicillium and Fusarium, are related to soil and plants. Higher occurrence in winter period could be explained by the agricultural surroundings and constructions in the areas near the facility. Alternaria spores had constant presence, during the whole study period, with higher counts during summer. Our observations are in accordance with observations made in a number of studies [8, 14, 15]. Table 3. Seasonal variation of eight airborne fungi in outdoor air and in air from both indoor locations. Mean Fungal Load () and entage () of total appeared. Autumn 2004 Winter 2004/05 Spring 2005 Summer 2005 Outdoor -OUT1 Cladosporium sp Penicillium sp Aspergillus sp Alternaria sp Fusarium sp Mucor sp Yeasts sp Other* Indoor 1 IN1 Cladosporium sp Penicillium sp Aspergillus sp Alternaria sp Fusarium sp Mucor sp

7 Evagelos Beletsiotis et al. / Procedia Food Science 1 (2011) Autumn 2004 Winter 2004/05 Spring 2005 Summer 2005 Yeasts sp Other* Indoor 2 IN5 Cladosporium sp Penicillium sp Aspergillus sp Alternaria sp Fusarium sp Mucor sp Yeasts sp Other* *: NSF+Other fungi Indoor air profile showed a different fungal distribution. More specific, at IN5 location, Penicillium spp. were the predominant species, with percentages ranged between 43.8% (in autumn) and 57.7% (in summer), showing constant presence with minor fluctuations during the study period. Maximum Penicillium sp. load was observed in summer with CFU/m 3. Cladosporium sp. were the second in abundance fungal species identified, showing a stable contribution to the air fungal load, with values ranged between 23.4% (in winter) to 32.9% (in spring). Yeasts spp. had the maximum concentration and fungal load at the rainy months (autumn and winter) with 17.6% and 23.4% respectively. Same seasonal pattern appeared to have Aspergillus species albeit with very low air loads at 3.8 CFU/m 3 in autumn and 3.3 CFU/m 3 in winter period. The other airborne fungal species identified, were distributed evenly through the year, without statistical significant seasonality. In IN5 area, air quality in many aspects (percentage occurrence of fungal species, ratio of fungi loads) was alike outdoor. This observation is consistent with the hypothesis that outdoor air has an important influence on indoor air quality [14].Thus, outdoor comparison samples should be collected when indoor air quality investigations were conducted. On the other hand, differences observed in the air quality (fungal load, species occurrence) between interior and exterior of the facility, reflect the different microenvironment that exists inside the facility. Fungal dynamics inside the plant is markedly different from the outside. Fungal species occurring indoors have the ability to adapt to that environment (high humidity in air and surfaces and constant temperature around 18 o C). In the other inside area (IN1) studied, fungal profiles were similar to some notable differences compared to IN5. The highest air load was observed in spring (135.4 CFU/m 3 ) and the lowest in summer (3.3 CFU/m 3 ) after corrective measures were taken. The most abundant fungal species identified belonged to Penicillium sp. with highest occurrence in spring (72.1%, 97.6 CFU/m 3 ) and lowest in summer period with 28.6% contribution and low air load (1 CFU/m 3 ). Cladosporium sp. were the second most abundant species with the highest air values occurring in autumn (19.5 CFU/m 3 ) and the lowest in summer period at 0.2 CFU/m 3 (Table 3). Aspergillus sp. and Fusarium sp. reached the highest percentage values in autumn and winter period. The proportions of the other fungal species showed similarity in all seasons with the exception of Alternaria sp. and other fungi (non sporulating and other phyla) which showed an increase over summer period (11.4% and 34.1% respectively) albeit with very low air loads (bellow 1.2 CFU/m 3 ) in both cases. IN1 is part of the filling product area of the plant. Monitoring of the fungal load and species distribution during fall 2004 till spring 2005 inside the area, indicated that the airborne fungi occurred in

8 1058 Evagelos Beletsiotis et al. / Procedia Food Science 1 (2011) high levels and they constituted an important source of yoghurt contamination during processing and cup filling as our stress tests and consumers complaints indicated (data not shown). It was decided to install and operate in IN1 area HEPA filters class , designed to withhold fungal spores and as a result to reduce the fungal air load. These filters started to operate in the last week of May As a result, the airborne fungi load of IN1 location was reduced in summer, thirty times (from 93.6 to 3.5 CFU/m 3 ) and almost 100 times, when compared to the mean fungal load during summer period of IN5 area. The microbiological air quality achieved was comparable to Greek and France hospital units [13]; [9], and it is maintained in the same levels since then (data not shown). Monitoring of air quality, in terms of fungal spore abundance and occurrence, consists a crucial component of the plant HACCP study. Furthermore, product contamination and customer complaints followed the fungal spore air reduction, resulting to a higher product quality and a significant microbiological safety. 4. Conclusions This study is unique because it describes culturable fungi obtained from a large number of indoor and outdoor air samples, during all four seasons of the year, in a Greek dairy plant. Analysis of the results allowed scientific conclusions and the decision of actions which led to product quality improvement. a. Fungal concentrations recorded in outdoor air were in a good accordance, when compared to other studies in Greece and other European countries. b. Fungal dynamics in both indoor areas studied (IN1, IN5) were very similar but markedly different from that observed in outdoor air. c. Cladosporium sp. were the dominant species outdoors and second in abundance indoors, while Penicillium sp. were the dominant species indoors. d. Seasonality of fungal genera was observed during the study. Outdoors lowest proportions of Cladosporium sp. were recorded during winter, while, in the same period of time, the higher proportions were obtained from yeasts and Penicillium sp. e. Indoor, the installation of HEPA filters in one of the two studied areas, significantly decreased the fungal load, almost 30 times, compared to load of airborne fungi retrieved before the installation. f. Implementation of constant monitoring of airborne fungal led to the acknowledgement of air quality as a critical point in a dairy plant. Acknowledgements We would like to thank Lecturer Dr. E. Kapsanaki-Gotsi and Dr. I. Pyri (University of Athens, Biology Dept.) for the preliminary air load studies and for valuable advices regarding technical issues. This research was in part supported by the General Secretariat for Research and Technology, Greek Ministry of Development and E.U (Project TP-5). References [1] Lourens-Hattingh, A. & Viljoen, B.C Yogurt as probiotic carrier food. International Dairy Journal, 11, [2] Cousin M.A Moulds in dairy products, In: Roginski H. (Eds.). Encyclopedia of Dairy Sciences. Oxford, Elsevier Science, USA. [3] Ottaviani F. & Ottaviani M.G Spoilage-Molds in spoilage, In: Caballero B. (Eds.), Encyclopedia of Food Sciences and Nutrition. Oxford, Academic Press, USA. [4] Kang, Y.J. & Frank J.F Biological aerosols: a review of airborne contamination and its measurement in dairy processing plants. Journal of Food Protection 52 (7),

9 Evagelos Beletsiotis et al. / Procedia Food Science 1 (2011) [5] Ren T.J. & Frank J.F A survey of four fluid milk processing plants for airborne contamination using various sampling methods. Journal of Food Protection 55 (1), [6] De Hoog G.S., Guarro, J., Gené, J. & Figueras, M.J Atlas of clinical fungi. Centraalbureau voor Schimmelcultures, Utrecht, The Netherlands, Universitat Rovira i Virgili, Reus, Spain. [7] Samson R.A., Hoekstra E.S., Frisvad J.C. & Filtenborg O Introduction to food-and airborne fungi. ASM Press, USA. [8] Pyrri I. & Kapsanaki-Gotsi E A comparative study on the airborne fungi in Athens, Greece, by viable and non-viable sampling methods. Aerobiologia 23, [9] Sautour M., Sixt N., Dalle F., L'Ollivier C., Fourquenet V. & Calinon C Profiles and seasonal distribution of airborne fungi in indoor and outdoor environments at a French hospital. Sci Total Environ 407, [10] Sveum W.H., Moberg L.J., Rude R. & Frank J.F. Microbiological monitoring of the food processing environment, In: Vanderzant C. & Splittstoeser D.F. (Eds). Compendium of Methods for the Microbiological Examination of Foods, 3 td. APHA, USA. [11] Ren T.J. & Frank F.J Sampling of microbial aerosols at varius locations in fluid milk and ice cream plants. Journal of Food Protection 55 (3), [12] Salustiano C.V., Andrade J.N., Brandao S.C., Azeredo R.M. & Lima S.A Microbiological air quality of processing areas in a dairy plant as evaluated by the sedimentation technique and a on stage air sampler. Brazilian Journal of Microbiology 34, [13] Panagopoulou P., Filioti J., Petrikkos G., Giakouppi P., Anatoliotaki M., Farmaki E., Kanta A., Apostolakou H., Avlami A., Samonis G. & Roilides E Environmental surveillance of filamentous fungi in three tertiary care hospitals in Greece. Journal of Hospital Infections 52, [14] Shelton B.G., Kirkland K.H., Flanders W.D. & Morris G.K Profiles of airborne fungi in buildings and outdoor environments in the United States. Appl. Environ. Microbiol. 68, [15] Horner E.W., Worthan A.G, & Morey P.R Air- and Dustborne Mycoflora in Houses Free of Water Damage and Fungal Growth Appl. Environ. Microbiol 70, Presented at ICEF11 (May 22-26, 2011 Athens, Greece) as paper MFS718.